|Publication number||US6859641 B2|
|Application number||US 09/886,516|
|Publication date||Feb 22, 2005|
|Filing date||Jun 21, 2001|
|Priority date||Jun 21, 2001|
|Also published as||US7228104, US20020197958, US20050159128|
|Publication number||09886516, 886516, US 6859641 B2, US 6859641B2, US-B2-6859641, US6859641 B2, US6859641B2|
|Inventors||Glenn D. Collins, Don L. Anair, Michael J. Ready|
|Original Assignee||Applied Signal Technology, Inc.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (33), Referenced by (60), Classifications (7), Legal Events (5)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This invention relates generally to communications systems, and more particularly to adaptive canceller architectures for use in multiple carrier frequency reuse communication systems, such as satellite communication systems.
In a conventional duplex communications system in which each of the sites of the system transmits and receives information to and from the other site, each site transmits on one frequency and receives information on another frequency. As a result, duplex communications require twice the bandwidth of a unidirectional communication link. This is costly where the available bandwidth is limited since it restricts the number of sites which can communicate over the system. This is particularly true with satellite communications systems. With satellites, transponder bandwidth is limited and system resources are costly.
In many relay-based communications systems, such as satellite links where the two transmitter/receiver sites will be in the same satellite beam, a downlink will be seen not only at the intended receiver site but also at the transmitter source of the downlink signal. Thus, each site receives not only the intended downlink signal from the other site, it also receives a replica of its own transmitted uplink signal. The replica comprises the uplink signal delayed in time by the roundtrip transmission time between the transmitter and the satellite, and it will be at a different signal level of course. Satellite and similar relay systems do not generally demodulate a received uplink signal, but merely translate, i.e., shift, the frequency of the uplink signal to a different frequency and transmit it back to the ground. The satellite on-board electronics that receives the uplink signal, frequency translates it to the downlink frequency, and transmits the downlink signal back to earth is called a transponder. Thus, the satellite transponder may receive uplink frequencies from two different sites in, e.g., the 6 GHz range, and translate the received signals to downlink frequencies in, e.g., the 4 GHz range. Each site could be assigned a different uplink frequency in the 6 GHz range and a different receive frequency in the 4 GHz range. As stated earlier, this requires twice the bandwidth of a unidirectional link.
Where both sites and, therefore, both the uplink and downlink signals are in the same satellite spot beam, systems have been developed that make more efficient use of bandwidth. These systems allow reuse of frequencies by two sites, thereby eliminating one uplink frequency and one downlink frequency. This doubles the available bandwidth of the satellite transponder. The basic idea behind these systems is that the received downlink signal at the first site is a composite signal comprising the sum of the desired signal of interest from the second site combined with a replica of the first site's own transmitted uplink signal. Thus, the signal of interest can be obtained by subtracting from the composite downlink signal the portion of the downlink signal due to the first site's own uplink signal. This requires generating a local replica of the uplink signal that has been relayed by the transponder. The relayed portion or replica of the uplink signal from the first site, received back on the downlink, is treated as interference and is removed from the composite downlink signal by an interference canceller.
In these known systems, the interference canceller is merely a device which produces and subtracts the replica of the interfering signal from the composite signal to give the signal of interest. The replica is derived from the uplink signal by estimating certain parametric changes imposed on the relayed uplink signal due to its roundtrip through the satellite transponder. These include not only the roundtrip time delay, but also the changes in signal level, frequency and phase that are experienced by the relayed uplink signal. These changes or parameters are estimated by the interference canceller and are applied to a local sample of the uplink signal as it is transmitted to produce the replica which is subtracted from the composite signal by the interference canceller.
The degree of cancellation of the interfering signal depends upon the accuracy with which the compensation parameters applied to the sample of the uplink signal can be estimated. Unfortunately, the changes experienced by the uplink signal are not static. Rather, they are dynamic and vary in time. Variations result from different causes. For one, satellites are not stationary, but tend to precess about a nominal location or position. This causes the distance between a transmitter/receiver site and the satellite to vary with time, which produces a varying time delay. It also produces variations in frequency and in phase due to Doppler shifts, and produces signal level variations. Moreover, the frequency translation circuits in the satellite transponder may also drift or vary in time, producing additional variations which are imposed upon the interfering downlink signal. Other effects that are relatively static but unknown include frequency-dependent amplitude and group delay distortion due to electronics at both the transmit and receive sites as well as in the satellite. The interference canceller must mimic these distortion components to provide adequate cancellation. Known interference cancellers attempt to estimate and track the static and dynamic variations in parameters to improve the accuracy of cancellation. However, this is a complex task. Known interference cancellers tend to be rather complicated, and lack the desired precision and accuracy in estimating the cancellation parameters to afford desired levels of cancellation of the interfering signal.
Additionally, known interference cancellers must be specifically designed or adapted for the satellite receiving systems with which they are intended to be used. This is because the interference canceller must be specially designed to interface to and be compatible with the satellite receiving system. This complicates the task of retrofitting or upgrading existing satellite receiving systems, or providing a universal interference canceller, since the interference canceller must be designed to be compatible with the electrical and mechanical specifications of the system hardware and software. Known interference cancellers also frequently must exploit a priori knowledge of certain external signal characteristics or the underlying data stream used to generate the uplink signal. This dependence also complicates any efforts to adapt the canceller to other systems.
It is desirable to provide adaptive interference cancellers which solve these and other problems of known interference cancellers by improving the compensation parameter estimation and cancellation functions while reducing complexity in the cancellers, affording a higher degree of precision and accuracy in canceling interfering signals (hence providing better performance), and leading to cost-reduced solutions. It is to these ends that the present invention is directed.
The invention solves the foregoing and other problems with present interference cancellers in a highly effective and efficient manner. The invention operates on IF signals prior to demodulation to cancel interfering signals by downconverting the signals to digital baseband signals, adaptively tracking and canceling time varying delay, amplitude and frequency offsets between the interfering signal and the local replica to generate a cancellation signal. The cancellation signal is subtracted from the received signal to produce the signal of interest. The signal of interest is then upconverted back to IF and demodulated in a normal manner using the conventional receiver electronics called a demodulator. The demodulator is typically included in the same electronics package with the modulator that generates the transmit signal. The combined modulator and demodulator is commonly referred to as a modem. This allows the invention to be a “drop in” element that can be readily used with existing receiving systems with little or no modification to the system.
In one aspect, the invention affords a method and system of canceling an interfering signal contained in a received signal which includes a signal of interest, where the interfering signal is a delayed and frequency offset version of a transmitted signal from the location of the received signal, in which the received signal and a sample of the transmitted signal are converted to digital form. Variations in delay and frequency offset between the interfering signal and the sample of the transmitted signal are tracked to produce a compensating signal which is combined with the received signal to cancel the interfering signal and produce the signal of interest. An error signal representing any uncanceled components of the interfering signal in the signal of interest is generated from the signal of interest, and used to control the tracking to minimize offsets between the compensating signal and the interfering signal.
In another aspect, the invention affords a method and system for canceling at a receiving site an interfering signal in a received signal that contains the interfering signal and a signal of interest, the interfering signal corresponding to a delayed and frequency offset version of a transmitted signal transmitted from the site, in which the received signal and a replica of the transmitted signal at intermediate frequencies are converted to digital baseband digital signals. Offsets corresponding to coarse delay and frequency differences between the transmitted signal and the interfering signal are applied to the replica of the transmitted signal to produce an initial compensated replica of the transmitted signal. Variations in delay and frequency offset between the initial compensated replica and the interfering signal are tracked to produce a second compensated signal which is combined with the received signal to cancel at least in part the interfering signal. Fine tracking and cancellation of any remaining portions of the interfering signal is accomplished by generating an error signal from the signal of interest and controlling the tracking with the error signal.
In yet another aspect, the invention affords an interference canceler for canceling an interfering signal corresponding to a delayed and frequency shifted version of a transmitted signal from a received signal containing the interfering signal and a signal of interest. The canceller comprises a variable delay for applying an initial delay compensation to a replica of the transmitted signal, and a variable frequency element for tracking changes in frequency and for applying an initial frequency compensation to the replica of the transmitted signal. An adaptive filter is employed for tracking delay, frequency dependent amplitude and phase variations following the initial compensations for delay and frequency to produce a resulting cancellation signal. The cancellation signal is combined with the received signal in a combiner to cancel the interfering signal and produce the signal of interest.
Other aspects of the invention will become apparent from the description which follows and as described below.
This invention is particularly applicable to satellite communications systems in which a satellite transponder translates the frequency of a received uplink signal to a new frequency and retransmits the received signal as a downlink signal on the new frequency, and will be described in that context. It will be appreciated, however, that this is illustrative of only one utility of the invention, and that the invention may be employed with any communication system which relays a received signal of interest as a composite signal that includes a replica of the receiving site's own transmitted signal. Examples include wireless and wireline communications systems.
The delay imposed upon an uplink signal from a given site comprises the time required for the transmitted signal to propagate from the transmit site to the transponder, the time delay experienced within the transponder during the frequency conversion process, and the transit time for the signal to propagate back down from the transponder to the receiving site. The transit time for the downlink signal to propagate back down to the site which transmitted it will be different from the downlink signal transit time to the other site since the sites are geographically separated. There are also other factors which produce signal variations. Since a satellite precesses about its nominal position in orbit, the time delay will be time varying due to the changing distances from the sites to the satellite. Moreover, the frequency offset, and the corresponding phase offset will also vary due to Doppler shift and drift in the electronics. Additionally, atmospheric effects such as fading, electronics in the uplink and downlink sites and changes in transponder characteristics will all produce time-varying changes in frequency, phase, signal level and time delay in the signal. These variations are illustrated diagrammatically in
Satellite transponders have limited bandwidth, typically about 36 to 72 MHz, which can be used to carry multi-channel communication signals such as FDMA, TDMA, or CDMA signals used for voice or data, as well as to carry television video and audio signals. Because the satellite transponder bandwidth is limited, and therefore a costly resource, it is desirable to reuse the available bandwidth for different purposes. Where transmit/receive sites A and B are within the same beam of the satellite transponder, and therefore receive a delayed and frequency offset replica of their own transmitted signal in the received downlink signal, systems have been developed which permit the two sites to reuse the same frequencies and bandwidth, for example, to afford a duplex communication link. This effectively doubles the available bandwidth of the transponder. These frequency reuse systems are based upon the principle that since the composite downlink signal comprises a linear combination of the signal of interest from the other site and an interfering signal comprising a delayed and frequency offset replica of the receiving site's own uplink signal, the interfering replica may be cancelled by locally generating an equivalent replica signal and subtracting it from the composite signal. These systems generate the local replica by estimating the delay and offset parameters and applying these to a sample of the uplink signal to produce a cancellation signal corresponding to the downlink replica. However, since the parameters are time varying, it is difficult to estimate or track these compensation parameters accurately. Thus, known systems use rather complex signal processing techniques to estimate and track the changing parameters. These systems tend to be costly and may not achieve the desired degree of performance.
The adaptive cancellation system 20 of the invention avoids complexity and affords a highly effective cancellation system which is easily implemented in existing communications systems. As will be described in more detail shortly, the cancellation system preferably operates at an intermediate frequency (“IF”) (typically 70 MHz) between the downlink downconverter 40 and uplink upconverter 42 and the satellite transmit and receive modem 44. This is advantageous since the adaptive cancellation system may be inserted into the IF of the receiver before demodulation, thereby obviating any modification to the modem or other processing circuits of the satellite receiving or transmitting system. Since it operates at IF, the adaptive cancellation system of the invention may be easily interfaced with existing receivers. Moreover, it requires no a priori knowledge of the upstream system. Thus, the canceller may be a “drop-in” component which may be used with existing satellite receiver systems.
As shown functionally in
As shown, the output of delay 102 may be supplied as one input to a multiplier, e.g., a mixer, 110, which has a second input from the output of a digital phase lock loop (DPLL) 112. The output of the multiplier 110 is supplied to a variable adaptive finite impulse response (FIR) filter 120. The adaptive FIR filter output is supplied as one input to a combiner 122. The other input to the combiner is the downlink baseband signal which is the composite signal comprising the signal of interest and the interfering replica of the uplink signal. The output of the FIR filter, as will be described, comprises a compensation signal corresponding to the replica, but opposite in sign, i.e., 180° out of phase. The compensation signal is effectively subtracted from the downlink signal in the combiner 122 to provide as an output the signal of interest. The subtraction process cancels the interfering uplink signal.
The output of the FIR 120 is also supplied as one input to the digital phase lock loop (DPLL) 112, the other input of which is the composite downlink baseband signal 60 from the DDC 72 (FIG. 2). The digital phase lock loop serves to track the frequency variations of the downlink signal and to control the multiplier 110 to cause the frequency of the uplink sample used for the local replica to track the varying frequency of the downlink signal. The DPLL preferably comprises a digital implementation in the baseband digital microprocessor of a second order phase locked loop, in which the fundamental loop filtering operations are preferably performed at a decimated rate in order to reduce the processing load on the baseband digital microprocessor. The basic phase difference metric (common to the control of all phase locked loops) is derived from the two DPLL inputs by performing a complex cross correlation of the two inputs, i.e., complex conjugate multiplication followed by accumulation of the multiplication product samples, decimating (rate reducing) the correlation result to obtain a lower rate sample stream, and performing an arctangent (ARCTAN) operation to derive the desired phase. Subsequent loop filtering operations may be performed on the reduced rate phase data. By “decimating” is meant that rather than processing samples at the input data rate (that is, the input sampling and digitization rate) of the baseband digital processing system 80, samples are accumulated for a predetermined time, and the accumulated result is periodically sampled and processed. This results in a greatly reduced load on the microprocessor. At the output, the low rate, decimated and filtered phase data may be interpolated and smoothed to produce a high rate phase correction term for use by a numerically controlled oscillator (NCO), the digital equivalent of a local oscillator. This produces the desired output frequency correction term, which is in fact a complex exponential term at the original sample rate that may be used by the multiplier 110.
For proper cancellation, the estimate of the roundtrip delay of the uplink signal must be sufficiently accurate to fit within the time window of the adaptive FIR filter 120 of the canceller, as will be described, and the frequency offset between the composite signal A+B and the delayed version of the uplink signal must be small enough for the DPLL to acquire and track them. The delay and frequency estimation unit 106 provides a coarse initial estimate of the delay and frequency offset, and this is used to control the variable delay 46 and the variable frequency shift unit 48 (see
The delay and frequency estimation unit 106 may be a software process, running on the digital baseband microprocessor. From snapshots of the data (blocks of samples) on the two input channels (that is, the downlink 60 and uplink 62 channels), the delay and frequency estimation unit derives the differences in both delay and frequency of the uplink signal element which are common to both. These differences may be used to initialize both the delay unit 102 and the DPLL 112. The DPLL in particular requires that data at its inputs be fairly tightly aligned in both delay and frequency. The tight alignment is necessary in order for the DPLL to be able to subsequently acquire and continuously track the frequency difference of the uplink component common to both inputs. The delay and estimation process provides estimates which are close enough for the processing system 80 to acquire and track the differences in delay and frequency. They may be quite fine (as opposed to gross), and not require further refinement. The DPLL, for example, requires that its inputs be time aligned to about 1/(signal bandwidth), which is typically equivalent to about 2-3 time samples. The delay and estimation process, which will be described more fully below in connection with
The compensations provided by the variable delay 46 and the variable frequency shift 48 afford a degree of signal alignment which, in a static environment, would be almost sufficient to fully support the desired cancellation process. However, the satellite communications environment is, in fact, dynamic, and continual fine adjustments are required to maximize the cancellation. The DPLL 112 itself provides the fine adjustment and tracking of the frequency differences, while the fine delay, amplitude and phase adjustments are most efficiently handled via the adaptive FIR filter 120 rather than the delay 102 because the adaptive FIR filter can delay the signal in increments less than one sample by adjusting the filter group delay.
The initial delay and frequency estimation unit 106 may employ a known ambiguity processing algorithm, such as described by S. Stein, “Algorithms for Ambiguity Function Processing”, IEEE Transactions on Acoustics, Speech, and Signal Processing, Vol. ASSP-29, No. 3, pp 588-599, June 1981. This algorithm basically provides an efficient approach for performing a complex correlation between two input signals over a range of delays and frequencies. The result of such a computation is called an ambiguity function or an ambiguity surface. At the correct delay and frequency offset, a peak in the magnitude of the ambiguity function will appear. This correlation is preferably implemented using discrete time fast Fourier transforms (FFT) for the calculation of the ambiguity function to minimize the computational requirements and, hence, the processing time, as will be described.
Delay 102 may be controlled by a delay tracking unit 108 which serves to control the delay imposed on the uplink baseband sample to track the changing satellite link delay produced by the satellite's precession. The delay 102 and delay tracking unit 108 operate using an error signal from an error generator/weight update unit 124.
The adaptive FIR filter 120 is controlled by the error generation and weight update unit 124. This unit, which will be described in more detail in connection with
The arrangement of the phase lock loop 112 as shown in
MOP serves as an error criterion which determines when, by how much, and in what direction the output of the adaptive filter is in error. It is a forcing function for driving the adaptive FIR filter in the right direction. LMS is used as a preferred algorithm for actually updating the filter taps or weights in response to an error that has been generated by the MOP error criterion though other adaptive weight update algorithms may be used. In order to achieve the maximum possible cancellation from the combining process, which mathematically is just an addition of the two signals, the signals must be matched exactly in phase (ignoring for the moment the requisite 180° phase shift between them), amplitude, time delay, and frequency. The adaptive filter can adjust for and track relatively slowly varying changes in phase, amplitude and time delay. It can also track relatively small time varying phase changes, but it is limited in its capability to handle significant frequency offsets. To be more specific, the frequency offsets it can handle are generally much less than the range of offsets encountered in practice. Thus, the DPLL function is required to reduce the frequency differences between the two signals to a point where they have negligible impact on the cancellation process. Thus, the DPLL function really works as a compliment to the adaptive filter in that it tracks out the frequency differences, while the adaptive filter tracks out amplitude, phase, and delay differences.
As shown in
The three taps 154 of the filter which are normally used are shifted among the various stages, i.e., delays of the filter, so as to maintain the center tap at the center of the filter. At a first time, n, the filter output is
y n =f 0 x n +f 1 x n−1 +f 2 x n−2.
At time n+1, the output of the filter becomes
y n+1 =f 0 x n+1 +f 1 x n +f 2 x n−1. (unshifted filter)
However, if it is decided that the centroid of the tap weights is no longer at f1 but has shifted instead towards f2, then the taps can be shifted such that f2 becomes the center tap. This can be accomplished by disconnecting f0 and connecting f3. The output at time n+1 will now be
y n+1 =f 1 x n +f 2 x n−1 +f 3 x n−2 (shifted filter)
Now the main tap is at the center of the filter as desired.
The filter operation described above is essentially a vector multiply, where the input vector is read from the delay memory and the filter vector is a set of coefficients. For normal operation with no shifts, the input data vector index is simply incremented for each new output. When a shift occurs in the filter taps, the input vector cannot be incremented as usual. In fact, in the example above, the same input samples used for generating yn are used for generating the shifted version of yn+1. This has an effect on the operation of the Read/Write address generator 142 (FIG. 5). When a shift occurs, the address generator may be adjusted to read out the correct input samples. As long as the delay memory has sufficient length to support the maximum delay of the system, a dual port RAM may be used for the delay memory, and is sufficient to implement the algorithm.
Cancellation begins at 178 with FIR adaptation. After a period of time (preferably of the order of 10 of the adaptive filter's time constants), delay tracking begins at 180, and the gain on the input to the digital up conversion process (82 in
The initial coarse processing determines an estimate of the relative delay and frequency offset between the uplink and downlink signals s1 and s2. In accordance with the invention, it does this by performing a series of complex cross correlations between the two signals s1 and s2. The correlations are performed over a wide range of assumed offsets in both frequency and delay. Each correlation assumes a fixed delay and frequency offset within the search range, which may be set based upon a priori knowledge of offsets and delays from past experience. The assumed delay and frequency offset are used to “adjust” one of the two inputs prior to performing the correlation. The correlation process is then performed, a different assumed offset is imposed, and the correlation process is repeated. The collection of results from many such correlation processes form a complex correlation “surface”. Each element of this surface corresponds to the correlation for a unique delay/frequency offset pair. When the input signals are closely matched, i.e., highly correlated, with one another, the magnitude of the correlation surface will exhibit a strong localized peak. The peak (in magnitude) may be easily identified because it rises above the noise-like surface level that is characteristic of the other delay and frequency-offset pairs of the correlation surface. The location of the peak identifies the delay/frequency-offset pair which maximizes the cross correlation between the two signals, and hence provides an initial coarse estimate of these differences. The previously referenced Stein IEEE publication gives insights as to how one may configure initial search parameters, and indicates why such a process provides coarse estimates of delay and frequency.
The invention implements the coarse delay/frequency offset process in a highly advantageous manner so as to minimize the computational time and complexity otherwise required by exploiting certain properties of the Fourier transform to efficiently compute the requisite complex cross correlations. By implementing a correlation in the frequency domain, using the well-known product rule for transforms, correlations can in effect be performed in parallel for a range of delays corresponding to the extent of the input signal records used to compute the Fourier transforms. This is illustrated in FIG. 8.
As shown, the two complex input signals may each be snapshotted (acquired as blocks of samples) and Fourier transformed at 200 using an FFT process. A complex element-by-element product (202) of the two frequency domain sequences is then formed after first conjugating and shifting one of the two inputs at 203, and the result (204) is inverse Fourier transformed at 206 using an IFFT process. The resulting sequence at 208 represents the complex cross correlation of the two input signals.
The relative correlation delays of the FFT based cross correlation process span the range −N/2 to +N/2 where N is the duration of each input record or sample block. All are preferably computed in parallel by the FFT-based cross correlation. It is desirable to limit the search range to half of this range in order to minimize distortion in the output, which may result from the cyclic nature of a convolution when implemented with an FFT. This is a phenomena well-known to those skilled in the art and discussed in many texts on digital signal processing. The distortion is eliminated by constraining the data in one of the two input signals to a length of N/2, instead of N, and “zero padding” the rest of the input data values in the length N input vector.
Similarly, the invention exploits another property of Fourier transforms to efficiently generate the relative frequency offsets required. Relative frequency offsets between the two inputs can be implemented by pre-multiplying one of the two signals prior to the FFT process. However, this represents an added computational burden. Instead, the process employed by the invention implements an effective frequency shift of the data by shifting the position within the output array of the FFT. Computationally this is much more efficient than the equivalent pre-multiplication operation.
The product at 204 of the two transformed (FFT) sequences, one of which has been conjugated and shifted to implement the frequency offset, represents the frequency domain equivalent of a cross correlation between the original inputs. Inverse transforming this with an IFFT operation at 206 produces the desired cross correlation at 208.
The cross correlation outputs are prepared for further processing by performing an absolute value operation at 210. This is necessary because the magnitude of the correlation operation is required for subsequent processing. It is the magnitude, not the phase, which determines the degree to which the two signals are correlated.
Following the absolute value operation, the individual elements of the correlation (magnitude) are compared at 212, 214 with a threshold to identify candidate correlation peaks. The success of the technique is dependent upon a judicious choice of threshold. The invention employs a process wherein an estimate of the standard deviation for the noise like (correlation free) segment of the cross correlation output is computed. This is computed by forming the cross correlation using inputs with sufficient separation in delay and frequency offset to guarantee that they are uncorrelated. The threshold may then be set proportional to the standard deviation estimate. Using a threshold derived from the data itself provides much better sensitivity and increased reliability over a fixed absolute threshold. When the foregoing processing has identified a correlation peak, it becomes a candidate for confirmation as a real peak and for further processing (shown in
The complex cross correlation of the two input signals s1 an s2 is implemented by performing a complex conjugate multiplication at 220 of frequency offset (222) and delayed, i.e., time shifted, (224) versions of the two inputs, and accumulating, i.e., integrating, at 226 the result. The input signal s1 may be frequency offset by pre-multiplying it at 222 with a complex exponential function having the appropriate frequency. The input signal s2 is delayed at 224 by shifting it ±1 sample.
The path for confirmation of a candidate peak is somewhat similar to the path for coarse correlation peak detection in FIG. 8. The magnitude of the correlation function may be computed with an absolute value at 230, and compared at 232 with the same threshold that was previously used for the coarse mode threshold detection. In the confirmation process, however, the correlation magnitude values are preferably averaged over N successive values at 234, prior to the threshold comparison, to minimize the likelihood of falsely detected peaks from the initial coarse mode process. The process may be repeated for points within a small neighborhood about the initial peak.
Confirmation of a peak may then initiate a fine mode frequency analysis process. The one sample resolution in delay associated with the coarse and confirmation modes is sufficient for subsequent processing, and hence there is no need for further refinement of the delay estimate.
The basis for the fine mode frequency estimate is a set of N complex correlation values. This is the same set of N values whose averaged magnitudes confirmed the selected final peak, which were stored at 240. An N point FFT is performed at 242 on the N values. N has preferably been selected to be a power of 2 to facilitate performing an FFT at 242. The absolute value of the outputs from the FFT are computed at 224, and the peak value determined at 246. The location (index) of the peak value within the output FFT array may be scaled to provide a fine frequency estimate at 248.
While the foregoing has been with respect to preferred embodiments of the invention, it will be appreciate by those skilled in the art that changes may be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US3419682||Jan 17, 1966||Dec 31, 1968||Northern Electric Co||Long transit time transmission system|
|US3750179||Nov 8, 1971||Jul 31, 1973||Bendix Corp||Single channel duplex space linked pulse communications system|
|US3993867||Oct 15, 1974||Nov 23, 1976||Motorola, Inc.||Digital single signal line full duplex method and apparatus|
|US4086534||Feb 14, 1977||Apr 25, 1978||Network Systems Corporation||Circuit for wire transmission of high frequency data communication pulse signals|
|US4140972||Apr 11, 1977||Feb 20, 1979||Nasa||System for synchronizing synthesizers of communication systems|
|US4146838||Sep 6, 1977||Mar 27, 1979||Nippon Electric Company Ltd.||System for detecting by a first pilot and a group of second pilots correlated to the first a frequency and/or phase difference between relayed carriers|
|US4178550||Jun 3, 1977||Dec 11, 1979||Bell Telephone Laboratories, Incorporated||Method and apparatus to permit substantial cancellation of interference between a received first and second signal|
|US4320498||Feb 11, 1980||Mar 16, 1982||Apple Computer, Inc.||Auto balancing duplexer for communication lines|
|US4385378||Aug 25, 1980||May 24, 1983||Communication Satellite Corporation||High power multiplexer for dual polarized frequency reuse earth stations|
|US4393494||Oct 2, 1980||Jul 12, 1983||Cselt Centro Studi E Laboratori Telecomunicazioni S.P.A.||Transceiver for full-duplex transmission of digital signals over a common line|
|US4439769 *||Oct 2, 1975||Mar 27, 1984||The United States Of America As Represented By The Secretary Of The Air Force||Combined adaptive sidelobe canceller and frequency filter system|
|US4470138||Nov 4, 1982||Sep 4, 1984||The United States Of America As Represented By The Secretary Of The Army||Non-orthogonal mobile subscriber multiple access system|
|US4516236||Sep 26, 1983||May 7, 1985||Northern Telecom Limited||Full-duplex transmission of bit streams serially and in bit-synchronism on a bus between two terminals.|
|US4535445||Jun 16, 1983||Aug 13, 1985||At&T Information Systems||Conferencing system adaptive signal conditioner|
|US4593399||May 9, 1983||Jun 3, 1986||Racal Data Communications Inc.||Method and apparatus for maximizing efficiency in satellite communications|
|US4670885 *||Feb 26, 1985||Jun 2, 1987||Signatron, Inc.||Spread spectrum adaptive antenna interference canceller|
|US4701935||Jan 9, 1986||Oct 20, 1987||Nec Corporation||One frequency repeater for a digital microwave radio system with cancellation of transmitter-to-receiver interference|
|US4712235||Nov 19, 1984||Dec 8, 1987||International Business Machines Corporation||Method and apparatus for improved control and time sharing of an echo canceller|
|US4896371||Nov 20, 1987||Jan 23, 1990||Kahn Leonard R||Synchronous AM transmission system having reduced self interference effects|
|US4952193||Mar 2, 1989||Aug 28, 1990||American Nucleonics Corporation||Interference cancelling system and method|
|US4989262 *||Nov 27, 1989||Jan 29, 1991||Nec Corporation||Demodulator with multiple interference cancellers responsive to correlations between undesired signals and error signals|
|US4995104||May 8, 1989||Feb 19, 1991||At&T Bell Laboratories||Interference cancelling circuit and method|
|US5131032||Mar 13, 1990||Jul 14, 1992||Hitachi, Ltd.||Echo canceller and communication apparatus employing the same|
|US5193112||Jan 24, 1991||Mar 9, 1993||Nec Corporation||Arrangement and method for rapidly detecting double talk in long distance telephone communications system|
|US5307405||Sep 25, 1992||Apr 26, 1994||Qualcomm Incorporated||Network echo canceller|
|US5436887||Jan 29, 1993||Jul 25, 1995||Storage Technology Corporation||Digital full-duplex transceiver|
|US5444864||Dec 22, 1992||Aug 22, 1995||E-Systems, Inc.||Method and apparatus for cancelling in-band energy leakage from transmitter to receiver|
|US5450456||Nov 12, 1993||Sep 12, 1995||Daimler Benz Ag||Method and arrangement for measuring the carrier frequency deviation in a multi-channel transmission system|
|US5475731||Jan 7, 1994||Dec 12, 1995||Ericsson Inc.||Echo-canceling system and method using echo estimate to modify error signal|
|US5596439||Aug 1, 1995||Jan 21, 1997||Viasat, Inc.||Self-interference cancellation for two-party relayed communication|
|US5691978||Jan 21, 1997||Nov 25, 1997||Signal Science, Inc.||Self-cancelling full-duplex RF communication system|
|US5974301 *||Sep 18, 1996||Oct 26, 1999||Ludwig Kipp||Frequency cancelling system and method|
|US6011952||Jan 20, 1998||Jan 4, 2000||Viasat, Inc.||Self-interference cancellation for relayed communication networks|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7280626 *||Sep 6, 2006||Oct 9, 2007||Sirf Technology, Inc.||Synthesizing coherent correlation sums at one or multiple carrier frequencies using correlation sums calculated at a coarse set of frequencies|
|US7292547 *||May 22, 2002||Nov 6, 2007||The Directv Group, Inc.||Device and method for nodal multiple access into communications channels|
|US7317757 *||Jul 22, 2005||Jan 8, 2008||Gennum Corporation||Transmit amplitude independent adaptive equalizer|
|US7340213 *||Jul 14, 2004||Mar 4, 2008||Atc Technologies, Llc||Intra- and/or inter-system interference reducing systems and methods for satellite communications systems|
|US7415064||Jul 22, 2005||Aug 19, 2008||Gennum Corporation||Transmit amplitude independent adaptive equalizer|
|US7418248||Apr 22, 2005||Aug 26, 2008||Ntt Docomo, Inc.||Signal reception device, signal transmission device, radio communication system, and signal reception method|
|US7430257||Apr 24, 2002||Sep 30, 2008||Lot 41 Acquisition Foundation, Llc||Multicarrier sub-layer for direct sequence channel and multiple-access coding|
|US7522877||Aug 1, 2008||Apr 21, 2009||Emc Satcom Technologies, Inc.||Noise reduction system and method thereof|
|US7593449||Jan 8, 2007||Sep 22, 2009||Steve Shattil||Multicarrier sub-layer for direct sequence channel and multiple-access coding|
|US7729657||Jul 7, 2006||Jun 1, 2010||Abel Avellan||Noise reduction system and method thereof|
|US7796986 *||Sep 14, 2010||Atc Technologies, Llc||Modification of transmission values to compensate for interference in a satellite down-link communications|
|US7860200||Oct 12, 2007||Dec 28, 2010||Harris Corporation||Communications system using adaptive filter that is selected based on output power|
|US7864835||Oct 12, 2007||Jan 4, 2011||Harris Corporation||Communications system using adaptive filter and variable delay before adaptive filter taps|
|US7941091 *||Jun 14, 2007||May 10, 2011||Rf Magic, Inc.||Signal distribution system employing a multi-stage signal combiner network|
|US7965761||Dec 5, 2008||Jun 21, 2011||Lot 41 Acquisition Foundation, Llc||Multicarrier sub-layer for direct sequence channel and multiple-access coding|
|US8081722||Dec 20, 2011||Harris Corporation||Communications system and device using simultaneous wideband and in-band narrowband operation and related method|
|US8085707||Sep 27, 2007||Dec 27, 2011||The Directv Group, Inc.||Device and method for nodal multiple access into communications channels|
|US8094763||Jan 10, 2012||Harris Corporation||Communications system using adaptive filter with adaptive update gain|
|US8098781||Jan 17, 2012||Harris Corporation||Communications system using adaptive filter with normalization circuit|
|US8107572||Jan 31, 2012||Harris Corporation||Communications system using adaptive filter for interference reduction|
|US8116419 *||Jul 14, 2008||Feb 14, 2012||Alcatel Lucent||Methods and apparatuses for estimating time delay and frequency offset in single frequency networks|
|US8121236||Oct 12, 2007||Feb 21, 2012||Harris Corporation||Communications system using adaptive filter circuit using parallel adaptive filters|
|US8204164||Jun 19, 2012||Harris Corporation||Communications system using adaptive filter and selected adaptive filter taps|
|US8238817||Aug 7, 2012||Emc Satcom Technologies, Llc||Noise reduction system and method thereof|
|US8364184||Sep 4, 2009||Jan 29, 2013||Gilat Satellite Networks Ltd.||Signal cancellation in a satellite communication system|
|US8374564||Aug 2, 2011||Feb 12, 2013||Comtech Ef Data Corp.||Signal filtering system and related methods|
|US8428198 *||Apr 23, 2013||Qualcomm Incorporated||Frequency tracking which adapts to timing synchronization|
|US8553805||May 5, 2011||Oct 8, 2013||Peter Monsen||Method and apparatus for subband signal demodulation in a transponder satellite communication link containing a component of relayed interference|
|US8644866||Feb 20, 2011||Feb 4, 2014||Novelsat Ltd.||Satellite receiver with interfering signal cancellation|
|US8654908 *||Jun 13, 2011||Feb 18, 2014||Comtech Ef Data Corp.||Correlation prevention methods for satellite adaptive cancellation links|
|US8655282 *||Oct 24, 2011||Feb 18, 2014||Qualcomm Incorporated||Multiple signal transformation in wireless receivers|
|US8670705||Jan 14, 2005||Mar 11, 2014||Atc Technologies, Llc||Additional intra-and/or inter-system interference reducing systems and methods for satellite communications systems|
|US8705435 *||Dec 21, 2011||Apr 22, 2014||The Directv Group, Inc.||Device and method for nodal multiple access into communications channels|
|US8774731 *||Jun 19, 2009||Jul 8, 2014||Astrium Limited||Compensation apparatus|
|US8837627 *||Feb 18, 2014||Sep 16, 2014||Comtech Ef Data Corp.||Correlation prevention methods for satellite adaptive cancellation links|
|US8855692||Dec 21, 2012||Oct 7, 2014||Gilat Satellite Networks, Ltd.||Signal cancellation in a satellite communication system|
|US9215019||Aug 2, 2012||Dec 15, 2015||Alcatel Lucent||Method and apparatus for interference cancellation in hybrid satellite-terrestrial network|
|US20050037749 *||Jul 14, 2004||Feb 17, 2005||Karabinis Peter D.||Intra-and/or inter-system interference reducing systems and methods for satellite communications systems|
|US20050136836 *||Jan 14, 2005||Jun 23, 2005||Karabinis Peter D.||Additional intra-and/or inter-system interference reducing systems and methods for satellite communications systems|
|US20050245220 *||Apr 22, 2005||Nov 3, 2005||Ntt Docomo, Inc.||Signal reception device, signal transmission device, radio communication system, and signal reception method|
|US20050254565 *||Jul 22, 2005||Nov 17, 2005||Apu Sivadas||Transmit amplitude independent adaptive equalizer|
|US20050254567 *||Jul 22, 2005||Nov 17, 2005||Apu Sivadas||Transmit amplitude independent adaptive equalizer|
|US20060217070 *||Mar 3, 2006||Sep 28, 2006||Atc Technologies, Llc||Modification of transmission values to compensate for interference in a satellite down-link communications|
|US20070002978 *||Sep 6, 2006||Jan 4, 2007||Anant Sahai||Synthesizing coherent correlation sums at one or multiple carrier frequencies using correlation sums calculated at a coarse set of frequencies|
|US20070211786 *||Jan 8, 2007||Sep 13, 2007||Steve Shattil||Multicarrier Sub-Layer for Direct Sequence Channel and Multiple-Access Coding|
|US20070217525 *||Mar 15, 2006||Sep 20, 2007||Qualcomm Incorporated||Frequency tracking which adapts to timing synchronization|
|US20080089271 *||Sep 27, 2007||Apr 17, 2008||The Directv Group Inc.||Device and method for nodal multiple access into communications channels|
|US20090097539 *||Oct 12, 2007||Apr 16, 2009||Harris Corporation||Communications system using adaptive filter and variable delay before adaptive filter taps|
|US20090098828 *||Oct 12, 2007||Apr 16, 2009||Harris Corporation||Communications system using adaptive filter that is selected based on output power|
|US20090110033 *||Dec 5, 2008||Apr 30, 2009||Lot 41 Acquisition Foundation, Llc||Multicarrier sub-layer for direct sequence channel and multiple-access coding|
|US20100008458 *||Jan 14, 2010||Hong Jiang||Methods and apparatuses for estimating time delay and frequency offset in single frequency networks|
|US20100056192 *||Sep 4, 2009||Mar 4, 2010||Gilat Satellite Networks, Ltd.||Signal Cancellation in a Satellite Communication System|
|US20100220646 *||Sep 2, 2010||Astrium Limited||Compensation apparatus|
|US20110243260 *||Oct 6, 2011||Comtech Ef Data Corp.||Correlation Prevention Methods for Satellite Adaptive Cancellation Links|
|US20120108175 *||May 3, 2012||Qualcomm Incorporated||Multiple signal transformation in wireless receivers|
|US20140161204 *||Feb 18, 2014||Jun 12, 2014||Comtech Ef Data Corp.||Correlation Prevention Methods for Satellite Adaptive Cancellation Links|
|US20140340255 *||Apr 21, 2014||Nov 20, 2014||Dali Systems Co. Ltd.||Real-time locating system using gps time difference of arrival with digital off-air access units and remote units|
|US20150244450 *||Nov 4, 2013||Aug 27, 2015||Novelsat Ltd.||Echo cancellation in communication transceivers|
|WO2005034361A2 *||Jul 16, 2004||Apr 14, 2005||Atc Technologies, Llc||Intra- and/or inter-system interference reducing systems and methods for satellite communications systems|
|WO2005034361A3 *||Jul 16, 2004||Aug 4, 2005||Gary G Churan||Intra- and/or inter-system interference reducing systems and methods for satellite communications systems|
|U.S. Classification||455/63.1, 455/278.1, 455/139|
|International Classification||H04B17/40, H04B7/185|
|Jun 21, 2001||AS||Assignment|
Owner name: TRANSCENDENT TECHNOLOGIES, CALIFORNIA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:COLLINS, GLENN D.;ANAIR, DON L.;READY, MICHAEL J.;REEL/FRAME:011928/0968
Effective date: 20010618
|Sep 3, 2004||AS||Assignment|
Owner name: APPLIED SIGNAL TECHNOLOGY, INC., A CALIFORNIA CORP
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:TRANSCENDENT TECHNOLOGIES, INC., A CALIFORNIA CORPORATION;REEL/FRAME:015749/0422
Effective date: 20040810
|Aug 13, 2008||FPAY||Fee payment|
Year of fee payment: 4
|Feb 15, 2012||AS||Assignment|
Owner name: RAYTHEON APPLIED SIGNAL TECHNOLOGY, INC., CALIFORN
Free format text: CHANGE OF NAME;ASSIGNOR:APPLIED SIGNAL TECHNOLOGY, INC.;REEL/FRAME:027708/0958
Effective date: 20110131
|Jul 25, 2012||FPAY||Fee payment|
Year of fee payment: 8